Neil Weinberg FanGraphs Q&A – 9/3/14

Neil Weinberg: Hey all, happy Wednesday. I’ll be here at 3pm to chat about baseball and baseball stats. This means questions about sabermetrics, our data, and related FG-centered concepts are strongly encouraged.

Other things are welcome too, but as usual, I’m not good at fantasy and prospect stuff, so ask at your own risk.

I’m @NeilWeinberg44 on Twitter, so follow me there and ask me questions at other times too. I’ll hang til 4pm for sure, and until 5pm if you ask lots of questions.

Comment From Hunter
How do you think teams like the A’s quantify chemistry? How much of an impact do you think it has in wins?

Neil Weinberg: I don’t know. I’ve heard all sorts of ideas. I’m not really sure how to go about doing it. I think it probably makes a difference, maybe something like 0-5 wins per season? But we’re talking about all of those wins spread across 25+ players. I would imagine players do better when they are happy and good chemistry makes them happy. How to measure that? No idea. I won’t even pretend to suggest we could do it from the outside in any way

Comment From A Concerned Reader
Where’s Dave? Is he okay?

Neil Weinberg: Dave’s on vacation!

Neil Weinberg: Which is good for Dave and will finally give me a chance to get more chat views than he does! Although Kiley probably crushed me.

Comment From Chris
Looking at the incredible season that JD Martinez is having, what one stat stands out to you the most?

Neil Weinberg: ISO. You cannot succeed in MLB with a 5% walk rate and 25% strikeout rate unless you mash. He’s doing it now.

Comment From Hunter
What peripherals best indicate a change in a batter’s true talent level?

Neil Weinberg: Typically, I would look for something to do with pitch selection. Guys add power, but that’s harder to tell in a stat independent of other stuff. Plate discipline can really help and it’s much easier to see on the stat sheet

Comment From CH Smoot
I think it would be interesting to see what players’ longest hitless stretches are, e.g., (I’m making this up) “in May 1985 Spike Owen had an 0-for-26 stretch, and that was the longest hitless stretch of his career.” Then you could look at something like, “Of all MLB hitters with >1000 career AB, who has the shortest longest career hitless stretch?”

Neil Weinberg: Sounds like a fun post for someone to do somewhere.

Comment From Don Quixote
Can you think of another player historically who’s done what Alex Gordon has – developed strictly as an IF (or OF) & start his ML career there, then not only switch mid-career to the OF (or IF), but become an elite defender at that position?

Neil Weinberg: No one comes to mind, but I’m also kind of young. So I don’t know the development histories of a lot of older players. It’s rare though.

Comment From Pennsy
How large is the MOE on the final WAR stat? As in, if you put players x far apart in WAR you wouldn’t feel comfortable saying one actually had a better season than the other.

Neil Weinberg: It’s a little fluid, because we have more confidence in certain parts of WAR than others. But I would say if you are 2 wins above or below someone, I am virtually certain you are better or worse. Really anything greater than 1 win is my rule of thumb, but there are always exceptions.

So, 5.0 WAR and 4.8 WAR are the same.
5.0 WAR and 3.9 WAR are very likely different.
5.0 WAR and 2.9 WAR are almost certainly different.

Fair?

Comment From Pennsy
If Fangraphs were given a MLB club (and budget) to operate, but could only use the statistics it has at its disposal now for player development and management purposes, how far behind the rest of the league in information gathering would the site be?

Neil Weinberg: I think we could probably build a reasonable comparable stats department. We don’t have the expertise running a team and stuff, but I think you could take FG data and be fine. There’s only so much extra value in having HITf/x over ISO, wOBA, batted ball data, etc.

Comment From Pale Hose
Hey Neil. Did you get confirmation from Appleman on the wRAA park adjustment formula?

Neil Weinberg: Sorry about this! I wanted it to be in the Off release last week but I didn’t touch base with Appelman soon enough. Hoping to have this added soon. WAR update is coming in about two weeks, so it will be live before then!

Comment From Scott
I was very surprised to learn that fWAR for relievers includes a leverage component. Are there other components of WAR that aren’t covered on the library page (which I’m sure you’re hard at work updating?)

Neil Weinberg: Basically, the Library is missing stuff that is new since the original post went up. For example, we treat infield flies as equal to strikeouts in pitcher WAR. There was an announcement about this, but it didn’t get into the Library. Lots of little stuff, but that’s all going to change soon!

I’m hoping to have WAR updated to the point where no one can say “WAR? No one can calculate that!”

Neil Weinberg: I was talking to a writer from another outlet and told him I could write down position player WAR on a napkin. I hope I will be able to help all of you do the same soon

Comment From CuriousGeorge
why is there still no player page for Rusney Castillo?

Neil Weinberg: Has he played a game yet?

Comment From Rob
Hi Neil. Can you comment on which of these you prefer for evaluating a pitcher’s performance to date and (if different) for predictive value going forward: FIP, xFIP, and SIERA? Thank you.

Neil Weinberg: FIP tells you more about how the pitcher performed from a value sense, while xFIP and SIERA try to build in some regression to tell you more about their underlying level. I typically use FIP and xFIP. SIERA isn’t bad, but it gets pretty complicated quickly and isn’t that much better than xFIP.

Fun fact, I built something called xxxFIP but haven’t tested it’s predictive nature. It uses those xK, xBB, etc stats.

Comment From Rob
Has there been any thought to adding xBABIP (at least for the current season) to the player pages?

Neil Weinberg: I’m pretty sure this has been discussed, but I don’t know if there’s consensus on its implementation.

Comment From Matt
I am looking for a way to download per game WPA data for pitchers. Is there a way to do it on Fangraphs?

Neil Weinberg: Per game: http://www.fangraphs.com/st…

Per play: http://www.fangraphs.com/st…

Comment From Hank G.
You (Fangraphs) post odds on all the games every day. For example, you think that the Nationals have a 65.2 % chance of winning today’s game with the Dodgers. Do you track the results of your predictions? If so, are they available somewhere?

Neil Weinberg: They aren’t published anywhere. I assume Appelman has them cataloged. They are new this year, so it wouldn’t shock me if Dave or someone did a retro post once the season is over.

Comment From Kiley McDaniel
Intimidated?

Neil Weinberg: Very.

Comment From Sean
I’m new(ish) to the sabremetrics society and just wanted to ask what the metrics say about Jose Reyes? Seems as though every time he steps into the plate or throws a runner out at first announcers rave about his tools but he just seems so much worse than perceived.

Neil Weinberg: Reyes is a terrific baserunner, decent defensively at SS, and a better than average hitter at his position. Earlier in his career, he was near MVP levels. Now he’s simply in the pretty good category. This happens with players as they age.

Comment From Lori
I know what WAR is, but I also see the term “fWAR” — what is that?

Neil Weinberg: fWAR is the notation we use to indicated that we’re talking about FanGraphs WAR. You might see bWAR or rWAR to talk about Baseball-Reference’s WAR.

Comment From A. Lane
Is there any anywhere that has balls in play broken down by percentage to different fields, and some type of pull%?

Neil Weinberg: If you go to splits, we have pull, center, oppo. You could calculate the percentage yourself. Trying to think if other sites have it already done? Not sure.

Comment From HookSlideBYB
Who do you think SHOULD win the AL Cy Young this year, and who do you think WILL win it? (Assuming they’re different)

Neil Weinberg: Would bet on Felix. Lester and Kluber have a chance. So do Price and Scherzer and Porcello if one of them go off.

I think I would support Felix today although I’m starting to think Sale has a real shot for my vote.

Comment From Mark
What kind of numbers can we reasonably expect from JD Martinez in the future? Marcus Thames-like numbers?

Neil Weinberg: I think .330-.340 wOBA with passable defense in a corner is plausible. He might be best as a 4th OF, but I’d let him play his way out of the lineup. The power is real.

Comment From Bill
I still relegate a lot of the Fangraphs pitching angle/batting stance change articles to the ghetto of ‘Spring Training/Changed His Approach”. Help me not do this.

Neil Weinberg: Spring Training is less meaningful because the competition isn’t putting up the same fight. If it happens during a season, you don’t have to worry that the pitcher was trying out a new pitch or something. Always want more data, but it’s not all small samples are equally useless.

Comment From Bomok
Do you guys get paid?o you do this out of the goodness of you hearts? Either way thank you vey much!

Neil Weinberg: Yup. All the FG writers are paid. Although, it’s not retire at 40 money!

Comment From Stew
What field position has the absolute least impact on a game? Consider all things equal in talent.

Neil Weinberg: Probably 1B. Although depends how you want to define “impact.” A horrible first baseman might ruin your infielders, but their ability to impact the game positively is limited relative to the rest

Comment From Pennsy
Got a favorite ballpark?

Neil Weinberg: Comerica, for biased reasons. PNC otherwise. That’s a freaking gem.

Comment From Jacob
Drew Smyly’s last start of the season is about 4-5 days away, what kind of leash do you expect the Rays to keep on him? Also do you put any amount of weight in the fact that he’ll be worth 1-1 1/4 fWAR with the Rays in 7 starts? What do you expect to see from him next year and things that the team will approach him about in the offseason?

Neil Weinberg: I think getting away from the Tigers outfield defense and moving to a better park is good, but also moving to good framers and a team that wants him to use his cutter more is good.

I think the Rays will let Smyly go as long as he feels comfortable. The 1+ WAR in 7 starts is a sign of what he could become. Still think he’s a back end starter unless he develops a good CH or split. Which we could. In which case he could be a #3?

Comment From Jacob
Do you think 1 or a mixture of Alex Colome/Mike Montgomery/Nate Karns would be enough to hold up until the perceived May return of Matt Moore?

Neil Weinberg: Yes. A month is nothing, as long as Moore is actually ready. Don’t want it to last three or four months with that crew

Comment From Guest
How are the Rest of the Season projections calculated? Does it take into account injuries and playing time?

Neil Weinberg: If you look in the FG Depth Charts, it’s ZIPS+Steamer/2 for the rate stats and then we manually predict playing time. On the player pages, it’s raw ZIPS or Steamer. Would have to ask them how they do playing time, but ours do.

Comment From Pale Hose
Neil – I was playing around with BABIP and wRC+ on Manny Machado’s 2013. Adding 10 points in BABIP (all singles) yielded roughly a 5 point increase in wRC+ and vice versa. I don’t know exactly what I was expecting, but I am surprised that wRC+ is that sensitive to BABIP. Thoughts?

Neil Weinberg: Haven’t done the calculation, but replacing outs with hits makes you better. 5% increase in offense is plausible.

Comment From Matt
Thanks! But is that the only way to do it? I’m looking for a way to download per game WPA data for multiple pitchers at once because doing that for every pitcher in a given season would take hours.

Neil Weinberg: I think you could probably use the play index at B-R for this?

Comment From Pale Hose
I mentioned this in Paul’s chat, but this seems like a good place to comment as well. Nate Jones faced 5 batters this year without recording an out. His ERA/FIP/xFIP numbers are calculated assuming he recorded 3 outs. Is this a glitch, or a simplifying assumption for pitchers that have 0 IP?

Neil Weinberg: I assume our system just can’t handle dividing by zero?

Comment From JJ
Assuming talent level is average, is there any advantage or disadvantage in having a left-handed closer? As an aside how overrated is the closer label/position?

Neil Weinberg: My guess is that it’s a disadvantage because more hitters are RHH, so you lose the platoon advantage. But I don’t know for sure how it plays out, given that a lefty closer is probably getting the job because they can get RHH out.

Very overrated. If I was in charge of a team, there wouldn’t be one.

Comment From Jheff
Did you know the Rangers use Fangraphs? I didnt know that, but thats what their Assistant GM said when he answered questions on Reddit.

Neil Weinberg: FG has lots of good stuff! Teams have good stuff too. I imagine that they use FG more to look at the rest of the league than at their own guys. Just a guess

Comment From Word
If evil forces made you evaluate batters by a single stat, and you could ONLY choose batting average or slugging percentage, which info would you rather have?

Neil Weinberg: SLG

Comment From Pennsy
NL MVP race is wide open. Using whatever standards you personally use, who makes your top five?

Neil Weinberg: Kershaw, Lucroy, Stanton, Cutch, Rendon? Feel good about those first three. Back end could go lots of ways.

Comment From Pale Hose
Math nerds want to see undefined instead of a made up denominator 🙂

Neil Weinberg: I know, but there are probably practical reasons for doing this. (RE Jones ERA)

Comment From Steve
Hey Neil, thanks for chatting! Are there any “new” sabermetric stats we can look forward in the near future?

Neil Weinberg: I will be curious to see the development of expected stats based on StatCast data. I have seen some of the expected stats based on batted ball and HITf/x stuff. Didn’t blow me away, but this data should be better.

Comment From Bomok
How valuable is Juan Lagares? This year he has one of the best WAR per plate appearences of any player in baseball

Neil Weinberg: His defense is amazing and he’s BABIPing his way to an average offensive season. I don’t think he’s an 7 win player for real, but saving runs in the OF is big, even if we want to regress his defense a bit.

Comment From Bomok
How accurate is this pitch framing study by baseball prospectus?http://www.baseballprospectus.com/article.php?articleid=16199

Neil Weinberg: I don’t really know. Max is very smart, so it’s probably as good as you can get given the data issues. Sorry I don’t have a more specific answer.

Comment From Word
Does fWAR *for pitchers* include batting/fielding performance?

Neil Weinberg: No. You have to go to their batting section to see that. Totally fine with me if you want to add that in when thinking MVP race, though, for example

Comment From John
Because an ace only pitches every 5 days, would it make sense to trade a true ace and fill your rotation with #2 & 3 talent starters while filling needs throughout the bullpen and lineup? I guess, how much value should we place on one pitcher in a rotation?

Comment From Matt
B/R’s play index only lets you download 300 records at a time. It would take maybe an hour straight to download a season’s worth of data.

Neil Weinberg: Really depends on what you get back. In theory it could work. Value is value, mostly

Neil Weinberg: An hour isn’t that long. I don’t mean to be snarky there. I find that if you do it by team it goes faster, but some of this stuff requires some elbow grease.

Comment From Guest
How often do you update the stats for the FG depth charts? Thanks for the chat

Neil Weinberg: The rate stats should be every night. The playing time varies based on who is editing that one and such.

Comment From like at wrigley?
do park factor calculations account for wind at particular times?

Neil Weinberg: No. Park factors account for the overall influence on run scoring over a number of years.

Comment From Brett
With today’s petty drama and the Sox’s frustration with JBJ, how much longer do you see him staying in Boston’s organization?

Neil Weinberg: I bet he gets traded this winter as part of a bigger deal.

Comment From Word
Does fWAR include batting/fielding performance?

Neil Weinberg: fWAR for position players includes batting, running, fielding, replacement level, and position

Comment From Jacob
Off the top of your head, are there any relievers or pitchers in AA/AAA that you see and think “Man, if someone got a hold of him and tinkered here and there..he’d be a steal”?

Neil Weinberg: Wrong person to ask. I think this about most minor league pitchers who don’t have a ++ fastball.

Comment From Guest
Why are all the components of a position player’s DRS listed except for catchers? I can get an infielder/outfielders DRS total by adding their rGDP, rARM, rGFP, and rPM together.

Neil Weinberg: Someone pointed this out recently, I haven’t had a chance to look at the BIS definition. Will try to have an answer for next week.

Comment From Guest
What does SLG and Bat measures of?

Neil Weinberg: Slugging is 1B+2*2B+3*3B+4*HR/AB. It’s a measure of bases per at bat, usually power.

Bat, in other data, is Batting Runs Above Average.

Comment From Oh No
what’s the best metric to use to explain to this Orioles friend of mine that his Os are living the dream but are way over performing their skill level?

Neil Weinberg: The problem is that the best stats for this are a bit complicated. Show him that the O’s happen to be performing better in high leverage moments (ie good timing).

He will assume this is a clutch skill.

then ask him why, if they have the ability to step it up in big moments, why they can’t do that in the 2nd inning?

Usually gets them

Comment From Rico Brogna
Is there research that measures the effectiveness of playing the infield in? As I type this, I’m pretty sure a soft ground ball just plated two in the 4th inning somewhere.

Neil Weinberg: Not sure if there’s a definitive study because we don’t track this publicly, but some estimates have been done and don’t make it look like a great idea.

Neil Weinberg: Quick break. Back in like two minutes. Keep the Qs coming.

Comment From jvetter
Is there anywhere I can calculate the league wide BABIP by hit type? IE – What is the BABIP in MLB in 2014 for ground balls, fly balls, etc.

Neil Weinberg: FanGraphs!

Neil Weinberg: http://www.fangraphs.com/le…

Comment From Word
Can you give a layman’s explanation of the difference between Total Zone and UZR?

Neil Weinberg: Total Zone uses play by play data from Retrosheet. It’s the best we can do for pre 2002 but UZR is just using much better inputs.

Comment From Me
What is the best defensive stat?

Neil Weinberg: UZR and DRS are both of equalish merit. I also like simple outs/balls in zone sometimes because it’s easy to communicate it to someone who isn’t as fluent.

Comment From Guest
Is there a way to take the stats found on the depth charts and convert them to SBs, HRs and other stats that my fantasy league values?

Neil Weinberg: Top of the page, click “Projections.” We have ZIPS and Steamer. Use one or the other, or find the averages (depth chart uses this).

Comment From Word
True or false? wOBA and wRC+ are batting stats exclusively, without baserunning included (except to the extent good baserunners stretch a double into a triple, etc).

Neil Weinberg: True. We used to include base running but do not do so anymore.

Comment From Word
Setting aside UZR and DRS, how big an upgrade is Total Zone over our fielding evaluation from 15 years ago, based on range factor, fielding%, and so forth?

Neil Weinberg: TZ is way better than fielding percentage, but it’s at the mercy of the data. Some games have more detailed info than others. Over a career, it’s way better. Over a single season, it’s better but it’s not perfect at all

Comment From Pennsy
Is positional flexibility in an of itself reflected in WAR? And how are the defensive adjustments split for calculating WAR if, for instance, a player shifts from 2B to 3B in the 7th inning?

Neil Weinberg: Positional adjustment gives credit based on defensive games played, which is scaled to the number of chances at each position and such. So it’s essentially going to be a fancy weighted average.

As for flexibility, there’s a case to be made that really flexible guys are a little undervalued from a roster building perspective

Comment From mtsw
Is it possible that good fielding adds value non-additively? i.e. having an elite 2B and SS playing in the field together may be cumulatively stronger in terms of limiting runs than having either of them alone? Or that having an entire defense full of above average guys could limit runs more than one of two elite defenders surrounded by replacement-level players even if the total UZR/DRS of the defenses are similar?

Neil Weinberg: Hmmm…I think so, if we’re counting positioning. Having an elite LF allows your CF to cheat to RF, things like that. Will have to think about it, but this is a very interesting question.

Comment From Brett
People keep saying that the Astros’ public image this year is “tarnished” – among other past-tense verbs – but with their player development, money saved and solid FO, does that really matter in either the short- or long-run?

Neil Weinberg: I think it’s bad to hurt your standing among players and agents, but I don’t think it’s nearly as big of a deal as some people are making it out to be. The Astros missed some chances to move forward, but they aren’t done for

Comment From Guest
Do you guys have a system to measure the accuracy of your projections? Not that I think you’re wrong or anything….

Neil Weinberg: Usually, at the end of the year, someone writes a post evaluating them. It’s very easy. Just go to the preseason projections and compare them to the 2014 stats!

Comment From CuriousGeorge
there seem to be an abnormal amount of super basic questions today

Neil Weinberg: Basic questions are good! Happy to answer them.

Comment From Jonah
There don’t seem to be many “true talent” stats for hitters – wRC+ I suppose is the closest you’ll get but that is still dependent on some BIP luck. Is there any metric you can think of that does a better job at determining true talent?

Neil Weinberg: Probably want to use some form of regression based wRC+. I’m sure this exists, but we basically use lean on using projections or last three years of data. They’re going to give you something close to true talent.

Comment From Chris
How much does the home team affect park factor? If the Rockies were a very poor hitting team, it would affect the park factor for Coors Field, wouldn’t it?

Neil Weinberg: No. Park factors are how well the home team and the road team perform in that park relative to all other parks. Bad Rockies are bad Rockies everywhere!

Comment From Steve
If I were to argue that pitching metrics are at a similarly imprecise level as defensive metrics, what would you say? Is there a legitimate argument that they are better?

Neil Weinberg: I would disagree. Defensive numbers are based on human inputs into a larger algorithm and those humans are using imperfect video of games. For pitchers, the location and speed of each batted ball is much less important than it is for fielding stats.

We undersell how imprecise hitting and pitching stats are, but they are better than fielding numbers

Comment From Brett
Any chance that FanGraphs adds minor league BIP & plate discipline stats? I feel like those are good groups of stats to look at when evaluating prospects.

Neil Weinberg: Don’t know if this is in the works, but have you been to Minor League Central? If not, see you in several hours.

Comment From Sad Sox Fan
Better 2015 Bogaerts or Betts?

Neil Weinberg: Hm…Bogaerts. But I like them both a lot.

Comment From Guest
Where can I find a breakdown of a hitters slugging% or ISO by direction (pulled, up the middle, oppo)? I know I’ve seen this somewhere.

Neil Weinberg: Splits. In the leaderboard, use the split drop down menu. In the player page click “splits” in the tab area. It’s there

Comment From mtsw
Any credence to the idea that improvement in O’s pitching is the result of Caleb Joseph/Nick Hundley being better framers/targets/game-callers? Improvement in pitching staff seems correlated roughly with Wieters’ injury

Neil Weinberg: Joseph rates as good framer and Hundley and Wieters look the same this year. I doubt this has been a huge factor, but I can’t really prove it either way off the top of my head

Comment From SimonSays
What do you think is the greatest unanswered question in sabermetrics right now?

Neil Weinberg: Honestly, how much credit to assign to the pitcher for the outcome of batted balls. FIP says 0, RA9 says 100%. It’s between those, but where?

Comment From Guest
I know there’s been some writing on FG earlier about this, but do you think the Mariner’s ultra-defense approach is sustainable? Would it make sense for a terrible offensive team like the Mets, who have a bunch of young pitching, to try and gather more Lagares type players?

Neil Weinberg: I don’t know if there’s research on this specifically, but saving a run and scoring a run are about equal. The key is to find great defenders who aren’t terrible hitters. That’s hard!

Comment From mtsw
Should analysts consider adding error bars or the like into projections? I feel like many fans now are currently misled into overrating minor league prospects and first-year players’ potential because they don’t understand how much uncertainty there is around how they’ll perform in the majors absent scouting context.

Neil Weinberg: Here’s the problem with that. The people who need the error bars are the people who are least fluent in interpreting statistics. Some projections publish these and everyone has them.

But people are really bad at understanding what a confidence interval is.

For example, Nate Silver said Obama was a 70/30 favorite to win in 2012. I bet 90% of the public couldn’t really explain that to you. It’s tough. But your point is valid

Comment From Guest
Thoughts on smaller market teams, i.e. Pirates. How much time is reasonable to see if there system works? If I was running the Pirates, I would scrap the current system and adapt what the cubs are doing.

Neil Weinberg: But the Pirates are good?

Comment From Scott
Re: the biggest unanswered question being pitchers and batted balls. Isn’t this what SIERA attempts to do, and seemingly does well since it performs better than any other ERA estimator? I’ve got to say I don’t really understand the reluctance to adopt it from FG.

Neil Weinberg: SIERA builds in assumptions that I’m not really convinced about. SIERA is trying to do what I’m saying. I’m not convinced it’s right.

Neil Weinberg: It barely beats xFIP despite it’s complexity. I would bet that if you ran a test that penalized for extra parameters, it would come out worse. But I’m not sure.

Comment From Word
“saving a run and scoring a run are about equal” … is this true? David Gassko wrote an article for HBT in 2006 saying a run saved is much better http://www.hardballtimes.co…

Neil Weinberg: Haven’t read that in a while, but I think that’s more of an issue at extremes. Which is why I said “about!”

Comment From Mike
Is FIP pronounced individually (F-I-P) or phonetically?

Neil Weinberg: I say “fip”, but apparently some people still say eff-eye-pee

Comment From Don
When/If hitF/X ever becomes public, could Tom Tango’s bbFIP become as reliable as wOBA?

Neil Weinberg: I don’t know. This is basically what I just said was the biggest unanswered question!

Comment From Paul
True or False: the increase in K league wise produces less balls in play, leading to even fewer chances for OF to be evaluated. Consequense: defensive metrics have larger error margins than before?

Neil Weinberg: Well, sort of. But the error bars are always conditioned on the sample size no matter what.

Comment From Scott
I think we should have a SIERA-specific chat some time since questions about it vs FIP seem to come up multiple times a week. I think for sabermetric laypeople it’s hard to grasp why it could do measurably (if only slightly) better at what it and xFIP both are trying to do yet get docked for nebulous other factors. What assumptions are built in that you’re not convinced about, and why should it be penalized for extra parameters?

Neil Weinberg: The correlation to Y2 ERA goes from like .38 to .41 from xFIP to SIERA.

xFIP has four parameters. SIERA has 9 I think. Statistically, adding more parameters will always increase your correlation. I’m not convinced the added model fit is substantive.

Comment From JW
Where did you guys get the template for your site? How much coding goes into the leaderboards (like, in hours)? I’m trying to do something similar for soccer, but as not a coder/web-designer I’m having a hard time determining how hard/much time these things take other people.

Neil Weinberg: I am not the right person to ask. I think it’s a pretty big undertaking to integrate all of this stuff but I’m an idiot regarding coding like this.

Comment From Larry Bernandez
Do you think the Indians next year should have Kipnis go to the OF, so that Ramirez can play full time 2b, and Lindor SS?

Neil Weinberg: I would stick with Kipnis at 2B for now. Let Ramirez push him off the position during a full season

Comment From Mike
minor feature request that I’ve made through the site before; if there’s a better way let me know. For those of us in NL-only or AL-only fantasy leagues, it would be nice if the minor league leaderboards could be filtered by league affiliation. You already have team affiliation so the data must be right there… (also filter on age would be nice too)

Neil Weinberg: Noted

Comment From Guest
Is the formula for BaseRuns the same for runs scored and runs allowed? I noticed on Tango’s website he lists a different formula for pitching statistics

Neil Weinberg: I think they are different.

Comment From Dylan
Jeff Luhnow tweet out today that Altuve should be considered for AL MVP. What are the chances that public defensive metrics are off enough that is a serious consideration? If we are wrong and defensive metrics warrents it, he is 1 in AVG,12th in OBP, and top 25 in wRC+

Neil Weinberg: Not buying it. Altuve has never graded very well on defense by our metrics. And he’s gotta make up two wins without the metrics also being wrong for Trout. Can’t see it. Top 10, sure?

Comment From mtsw
With increasing awareness of how important defense is, should we re-evaluate some past players? Presumably elite defensive skill was even more valuable in previous eras with lower strikeout ratios. Could Mark Belanger be a Hall of Famer even if we lack the tools to prove it now?

Neil Weinberg: Theoretically, you’re right. I’m sure there were good defenders who got totally overlooked and people who were very overrated. Not sure what we can do about it though

Comment From Matt
Today is my first day of Law School. How do I become Nick Faleris and somehow manage a to become a full time big shot lawyer and a expert amateur scout for a major baseball site who flies around the country and scouts high school kids?

Neil Weinberg: So as for the lawyer thing….no idea. As for the scout thing, read scouting stuff all the time. Watch tons of baseball. Watch a guy play and then guess what he’s good at and then check to see if you’re right.

Also, start writing about baseball very often.

if you can do that while in law school, you’ll do okay. I started writing during my first year in a PhD program

Comment From CuriousGeorge
who is tom tango?

Neil Weinberg: One author of The Book. Consults for at least one team, and blogs about saber stuff

Comment From Guest
I planned on getting a baseball related tattoo. I thought about some sabermetric terms being involved. Are there any metrics that won’t look stupid in 20 years?

Neil Weinberg: Great question.

Neil Weinberg: Hmmm….get your team’s logo. Or wRC+. That might look cool

Comment From Bryan Cole
I disagree with the reasoning behind your pitching/defense answer (above). I think sample size is a bigger issue than precision: even once the StatCast data are released, you’ll still only be judging defensive performance on the basis of the couple dozen balls on the edge of a player’s range, rather than the thousands of pitches a starter throws.

Neil Weinberg: Ha…I may have answered using accuracy instead of precision.

Comment From CuriousGeorge
in moneyball beane didnt care about defense. so today would a team try and play hatteberg at first?

Neil Weinberg: If he could get on base like crazy, sure. Defense is popular now because defense is cheaper than offense. Someday that will be different and Moneyball teams will chase something else

Comment From Brett
The D-Backs are thinking of retaining Kirk Gibson. After trashing their minors and the gritty/revenge stories, they should just clean house with him, his staff and Towers, right?

Neil Weinberg: Based on the information available to the public, with the realization that there may be info hidden from us, Towers and Gibson should be gone before the end of the World Series

Neil Weinberg: Get any last minute questions in!

Comment From Guest
Sadly, the more interested I become in baseball, and get into sabermetrics and scouting, the more I don’t have a team. I “root” for the Red Sox, but I have significantly less emotional attachement than a “fan” would. I like good baseball. If my team is no-hit, I just like that I saw a no-hitter. It seems to be common in the industry. See Dave and the MAriners.

Neil Weinberg: Yeah, this is common for a lot of people. But it’s also because we typically view baseball really objectively when we analyze it and it’s hard to remember to have fun when your team is on. I catch myself doing this a little. You can fight back.

Neil Weinberg: If you want

Comment From Ryan PM
Is there a valid argument for distributing the WAR positional adjustments more granularly? At some level, it seems odd that 2B and 3B would be measured against each other, and this year, Alex Gordon has clearly benefited from the fact that what is “average” defensively in left field is far inferior to what is average in right (even if it shouldn’t be that way).

Neil Weinberg: Yes. The positional adjustment could be more responsive to these things.

Neil Weinberg: Alright, let’s find a safe place to land.

Comment From Ryan PM
What is your favorite statistic that you’ve only become acquainted with within the last year or so?

Neil Weinberg: I didn’t really start using RE24 for relievers until recently. Does that count?

Comment From Brett
Mookie Betss: Great baseball name in the majors right now or greatest baseball name in the majors right now?

Neil Weinberg: Skye Bolt plays for UNC. Hope he gets drafted next year.

Neil Weinberg: Remember to check out the Library and the Glossary and everything.
“Def” is getting an entry this week. I’m on Twitter @NeilWeinberg44 and we’ll do this again next week

Neil Weinberg: Take care all